Method for optimizing the load on a main bearing of a wind turbine

Abstract

The invention relates to a method for optimizing the load on a main bearing (1) of a wind turbine, wherein the main bearing (1) is designed as a hydrodynamic plain bearing, in which at least one of the bearing rings (2, 3) consists of a number of bearing segments (4) which adjoin one another around the circumference of the bearing ring (2, 3), wherein each bearing segment (4) can be positioned in a defined radial position (r) and wherein each bearing segment (4) is arranged on a surrounding component (5). In order to achieve optimized operation of the bearing, the invention provides that the method comprises the steps of: a) carrying out a numerical simulation for each of the bearing segments (4) of the bearing ring (2, 3),in which the load on the bearing segment (4) is calculated when arranged on or in the surrounding component (5) when subjected to the expected bearing load and with a predetermined initial radial positioning (r0) of the bearing segment (4) relative to the surrounding component (5); b) carrying out a numerical simulation for each of the bearing segments (4) of the bearing ring (2, 3), in which the load on the bearing segment (4) is calculated when arranged on or in the surrounding component (5) when subjected to the expected bearing load and with a predetermined changed radial positioning (r1) of the bearing segment (4) relative to the surrounding component (5); c) comparing the loads on the bearing segments (4) determined according to steps a) and b) and selecting the radial positionings for all bearing segments (4),at which the most uniform possible load on all bearing segments (4) is to be expected; d) adjusting the radial positioning of all bearing segments (4) according to step c).

Description

Field of the InventionThe invention relates to a method for optimizing the load of a main bearing of a wind turbine, wherein the main bearing is designed as a hydrodynamic plain bearing, in which at least one of the bearing rings consists of a number of bearing segments which adjoin one another directly or indirectly via the circumference of the bearing ring (further components may also be arranged if appropriate between two segments), wherein each bearing segment can be positioned in a defined radial position and wherein each bearing segment is arranged on an environmental component.In addition to the classic main bearing of the rotor of a wind turbine by means of a rolling bearing, the use of a hydrodynamic sliding bearing is also possible. Such bearings are an advantageous alternative to rolling bearings in this respect because a bearing exchange on the tower of the wind turbine is possible without dismantling the rotor. Such a bearing can preferably have bearing rings segmented in the circumferential direction. This has the advantage, especially in the case of very large wind power installations, that no very large forging and hardening installations are required for producing the bearing parts and the transport of the bearing segments is much simpler.The individual bearing segments of the bearing ring must then be adjusted to their correct radial position during the assembly of the bearing in order to ensure the functionality of the bearing.The problem has been found here that the surrounding component, i.e. in particular the bearing housing, can be of different stiffness (measured in the radial direction) over the circumference of the bearing ring, which has the result that the individual bearing segments are loaded to different extents.SUMMARY OF THE INVENTIONThe object of the invention is therefore to propose a method of the type mentioned at the beginning for a segmented hydrodynamic plain bearing as a main bearing for a wind turbine, with which method it is possible to produce optimized operation of the bearing.The solution of this object by the invention provides that the method has the steps:a) carrying out a numerical simulation for each of the bearing segments of the bearing ring or a numerical simulation which takes into account all the bearing segments (4), in which the load of the bearing segment is calculated when it is arranged on or in the surrounding component when the expected bearing load is applied and when the bearing segment is initially positioned in a predetermined manner relative to the surrounding component;b) carrying out a numerical simulation for each of the bearing segments of the bearing ring, in which the load of the bearing segment is calculated when it is arranged on or in the surrounding component when the bearing load to be expected is applied and when the radial positioning of the bearing segment relative to the surrounding component is predefined and changed;c) comparing the loads of the bearing segments determined according to steps a) and b) and selecting the radial positionings for all bearing segments, in which the most uniform possible load of all bearing segments is to be expected;d) Setting the radial positioning of all bearing segments according to step c).Although even with this procedure, a homogenization of the load on the individual bearing segments can be achieved, a preferred embodiment of the proposed method provides that the step is carried out after step b above) and before step c above).b') Repeated performance of the numerical simulation for each of the bearing segments of the bearing ring according to step b above), wherein again changed radial positionings of the bearing segments are predefined;In this case, in the above step c), it is then provided that all the determined loads of the bearing segments are taken into account during the comparison and during the selection.The numerical simulation mentioned is preferably a finite element analysis (FEA).In the setting according to step d above), the individual bearing segments are preferably positioned radially in each case with at least one actuating element.The method is preferably carried out in the case of a hydrodynamic plain bearing whose outer ring has at least 6 bearing segments.Furthermore, it is preferably provided that the method is carried out only for the bearing segments of the bearing outer ring.If hydrodynamic plain bearings are used as main bearings in wind turbines, it is known that local load increases can occur in the region of relatively stiff locations of the housing (i.e. of the surrounding component). This results in a correspondingly higher load on the bearing, which is disadvantageous.For the geometric adjustment of the position of the individual bearing segments, i.e. for their radial positioning or height adjustment, adjusting elements are indeed known. However, until now only the geometric setpoint value of the bearing segment has been set, i.e. only a geometric adjustment of the bearing segments takes place.In order to advantageously reduce locally higher loads on bearing segments and thus to increase the service life of the bearing, the method according to the invention is used. In this way, in particular a longer service life can be achieved and thus the times for the replacement of bearing segments can be extended. Ideally, all bearing segments are evenly loaded, resulting in maximum useful life of the main bearing.The proposed method is based on the idea that the ambient rigidity of the housing and the bearing load to be expected are taken into account and all bearing segments are adjusted radially in such a way that they are loaded as uniformly as possible.By means of the finite element analysis (FEA) preferably used, the rigidity in the surrounding construction can be calculated in a manner known per se. The radial feed of the individual bearing segments, i.e. the height setting, can then be varied and the loads on the individual bearing segments resulting in each case can be determined.Thus, the optimum radial feed can be calculated iteratively or by using optimization algorithms known per se for each bearing segment and these can then be set. The aim here is for all bearing segments to be loaded as uniformly as possible. The result of the calculation is thus respective specification values for the specific height setting of each bearing segment. During the mounting of the bearing, these determined setting values are then used as the basis.Brief Description of the DrawingsThe drawings illustrate an embodiment of the invention. The following are shown: FIG. 1 schematically shows a main bearing of a wind turbine which is accommodated in a bearing housing, and FIG. 2 shows the detail "A" according to FIG. 1, with details given for the radial setting of a bearing segment.DETAILED DESCRIPTION OF THE DRAWINGSFIG. 1 outlines the main bearing 1 of a wind turbine which has an inner bearing ring 2 and an outer bearing ring 3. Both bearing rings 2, 3 are segmented, i.e. each bearing segment 4 (see FIG. 2 ) extends only over part of the circumference of the bearing ring; together, all bearing segments 4 then form the respective bearing ring. The main bearing 1 is disposed in a bearing housing 5 which constitutes an environmental component of the bearing.In the exemplary embodiment shown, both the bearing inner ring 2 and the bearing outer ring 3 each have 18 segments which adjoin one another in the circumferential direction.The outline shown in FIG. 1 reveals, by way of the shape of the bearing housing 5, that, with respect to the radial direction r, different stiffnesses are provided for the individual bearing segments by the bearing housing 5.FIG. 2 shows that the bearing segments 4 can be adjusted in the radial direction r (height adjustment), for which purpose adjusting elements 6 are used. These are, for example, a screw construction with which the bearing segment 4 can be fixed in a predetermined radial positioning relative to the bearing housing 5. The possibility of setting the bearing segment 4 by the adjusting elements 6 is indicated by arrows on the adjusting elements 6.The aim of the method described above for setting the individual bearing segments 4 is to position all bearing segments radially in such a way that, given a given bearing load, all bearing segments are loaded as uniformly as possible.Accordingly, it is first calculated by a finite element analysis (FEA) for an initial radial positioning r 0 which load the individual bearing segments have to bear, taking into account the local rigidity of the bearing housing 5 and the bearing load to be expected. This will initially regularly result, given an identical initial radial positioning for all bearing segments 4, in those bearing segments which are at stiff locations of the bearing housing being subjected to a higher load than those bearing elements which are at less stiff locations.The FEA can now be repeated with changed radial positionings, which is indicated in FIG. 2 with the positionings r 1, r 2 and r 3. Consequently, there are changed values for the load of the individual bearing segments 4.By varying the values for the radial positioning, optimized values for the individual radial positionings of the individual bearing segments 4 can now be determined. Optimization algorithms known per se can be used for this purpose.In principle, however, the method can also already be implemented in elementary terms in that a lower radial positioning is used as a basis for those bearing segments 4 which have to bear a higher load than the average over all bearing segments, and / or in that a higher radial positioning is used as a basis for those bearing segments 4 which have to bear a lower load than the average over all bearing segments.Thus, an adaptation of the load of all bearing segments can be determined iteratively by determining the respective radial positioning, which, after carrying out said calculation, are then used as a basis for the mounting of the main bearing.List of reference characters1 Main bearing 2 Bearing ring (bearing inner ring) 3 Bearing ring (bearing outer ring) 4 Bearing segment 5 Surrounding component (bearing housing) 6 Actuating element r Radial position of the bearing segment r 0 initial radial positioning r 1 changed radial positioning r 2 further changed radial positioning r 3 further changed radial positioning

Claims

Method for optimizing the load of a main bearing (1) of a wind turbine, wherein the main bearing (1) is designed as a hydrodynamic plain bearing, in which at least one of the bearing rings (2, 3) consists of a number of bearing segments (4) which adjoin one another directly or indirectly via the circumference of the bearing ring (2, 3), wherein each bearing segment (4) can be positioned in a defined radial position (r) and wherein each bearing segment (4) is arranged on an environmental component (5), characterized in that the method has the steps: a) carrying out a numerical simulation for each of the bearing segments (4) of the bearing ring (2, 3) or a numerical simulation which takes into account all the bearing segments (4), wherein the load of the bearing segment (4) is calculated when it is arranged on or in the surrounding component (5) when the bearing load to be expected is applied and when the initial radial positioning (r 0) of the bearing segment (4) is predefined relative to the surrounding component (5); b) a numerical simulation is carried out for each of the bearing segments (4) of the bearing ring (2, 3), wherein the load of the bearing segment (4) is calculated when it is arranged on or in the surrounding component (5) when the bearing load to be expected is applied and when the radial positioning (r 1) of the bearing segment (4) is predefined and is changed relative to the surrounding component (5); c) Comparison of the loads of the bearing segments (4) determined according to steps a) and b) and selection of the radial positionings for all bearing segments (4), in which the most uniform possible load of all bearing segments (4) is to be expected; d) Setting of the radial positioning of all bearing segments (4) according to step c).Method according to Claim 1, characterized in that after step b) and before step c) of Claim 1, the step is carried out: b') Repeated performance of the numerical simulation for each of the bearing segments (4) of the bearing ring (2, 3) according to step b) according to Claim 1, wherein changed radial positionings (r 2, r 3,...) of the bearing segments (4) are in turn predefined, wherein in step c) according to Claim 1 all the loads determined on the bearing segments (4) are then taken into account in the comparison and in the selection.Method according to claim 1 or 2, characterized in that the numerical simulation is a finite element analysis (FEM).Method according to one of Claims 1 to 3, characterized in that the bearing segment (4) is positioned radially with at least one actuating element (6) during the setting according to step d) of Claim 1.Method according to one of Claims 1 to 4, characterized in that it is carried out for a hydrodynamic sliding bearing whose outer ring (3) has at least 6 bearing segments.Method according to one of Claims 1 to 5, characterized in that it is carried out only for the bearing segments (4) of the bearing outer ring (3).

Images